Research On Motion Control Of Riding Lower Limb Assist Exoskeleton Robot

As an auxiliary device of helping human lower limb movement,the lower limb assisted exoskeleton robot is mainly involved with disciplines as control,mechanics,materials and bionics and others,which is widely used in many fields as military,medical,industrial,rescue and so on.In recent years,along with the social development and the aggravation of aging,the lower limb assisted exoskeleton robot has a wider market application prospect.In front of this opportunity,based on investigating the development situations of lower limb assisted exoskeleton robot in domestic and overseas,this thesis designs a lower limb assisted exoskeleton robot of riding type.It not only decreases the burden of human standing by riding,but also provides the auxiliary force for human lower limb when human moving,which can save more labor in moving.Except improving the happiness index of the elderly and frail groups,this thesis still has the key practical significance in solving social issues brought by insufficient resources and population aging.Based on the designed lower limb assisted exoskeleton robot of riding type,this thesis has refers to huge amounts of documents to carry out researching on key technologies and technological difficulties of lower limb assisted exoskeleton robot,and focus on the subject of motion control of lower limb assisted exoskeleton robot.The main tasks as below are completed during researching:（1）According to the biomechanics of human lower limbs and human anatomy,compared with physiological structure and motion characteristics of human lower limb,the brief analysis on mechanical structure of riding lower limb assisted exoskeleton robot are carried out to ensure basic parameters of exoskeleton robot.Through Solid Works software,the 3D mode of exoskeleton robot is obtained and the assistance type of exoskeleton robot is confirmed,which provides powerful theoretical support for building and analyzing mathematical model of exoskeleton robot.（2）According to the gait analysis of human walking,the single leg of riding lower limb assisted exoskeleton robot is simplified as a 5-link mechanism.The kinematic analysis is completed through building kinematics model of lower limb assisted exoskeleton robot by D-H parameter method.Based on kinematic analysis,the 5-link dynamic model of exoskeleton robot in single leg supporting stage is constructed by using lagrangian method and in accordance with this model,the dynamics analysis is completed,which provides theoretical basis for design of control rate and simulation verification of control algorithm.（3）Based on the kinematics and dynamic analysis of exoskeleton robot,a traditional fuzzy self-adaptive controller is designed,which compensates the uncertain factors as friction,loading change and external disturbance when the exoskeleton robot assists the human body.And it is verified that the algorithm can better realize the tracking control of human motion trajectory by simulation.（4）In the process of designing fuzzy self-adaptive controller,regarding the problems as difficulties of constructing adaptive rate function and limitation of self-adaptive ability,the improved grey wolf optimization of fuzzy self-adaptive control algorithm is brought up to optimize weight of fuzzy controller.In order to improve the efficiency and accuracy of optimization,two methods are introduced to optimize GWO（grey wolf optimization）,which are successfully applied fuzzy self-adaptive controller.Finally,through simulation,it is verified that optimized controller has advantages as faster convergence speed,smaller tracking error and stronger real-time performance than traditional one.（5）In order to do better researching on motion control problem of lower limb assisted exoskeleton robot,the controlling system of exoskeleton robot with master slave distributed architecture based on CAN bus is constructed.This controlling system is divided into 3 levels to conduct control strategies as decentralized control and centralized management.Based on this architecture,and according to the control demands of exoskeleton robot,the relevant designs of controlling system as main controller,drive module,acquisition module and power module are respectively improved,which provides good basis for sample construction of exoskeleton robot.Finally,in the Adams dynamics simulation software,the riding simulation of riding lower limb assisted exoskeleton robot studied in this thesis is carried out.It is proved that it can track human movement very well and provide force for human lower limbs.The exoskeleton robot studied in this thesis is mainly for the elderly and groups who need to stand in a long time.Through above tasks,it is shown that the control algorithm and control system designed in this thesis realize the design target that the exoskeleton robot can provide effective force for human lower limb,which lay a foundation for further research in the future.